MARKET WHEELS

MKTG - LEE

Executive Summary

This project analyzes the used car market to provide actionable insights for dealerships and automotive marketers. Using a dataset of 98,836 vehicle listings across 9 major brands (Mercedes, BMW, Audi, VW, Toyota, Ford, Hyundai, Skoda, and Vauxhall), we examined pricing optimization, market segmentation, and brand positioning strategies.

The primary challenge facing automotive dealerships is identifying which vehicle attributes consumers find most desirable and understanding how to strategically price and position inventory to maximize profitability while meeting diverse customer needs.

Through conducting exploratory data analysis, K-means clustering, PCA visualization, logistic regression, and multiple linear regression, we identified key market trends, customer segments, and brand positioning opportunities that can guide inventory acquisition, pricing decisions, and targeted marketing strategies.

Problem Statement

The used car market presents significant challenges for dealerships seeking to optimize their operations and marketing strategies. With thousands of vehicles varying across brand, age, mileage, engine specifications, and fuel type, dealerships struggle to make data-driven decisions about three critical areas:

1.Pricing Optimization

How should dealerships price their inventory to remain competitive while maximizing profit margins? Which vehicle attributes have the strongest influence on price, and how can this knowledge inform acquisition and pricing strategies?

2.Market Segmentation

Who are the target customers in the used car market? Can distinct buyer segments be identified based on vehicle preferences, and how should marketing efforts be tailored to reach each segment effectively?

3.Brand Positioning

How do automotive brands compare in terms of perceived value and market positioning? Which brands command premium pricing, and where do opportunities exist for competitive differentiation?

This project addresses these challenges by leveraging a dataset of 98,836 vehicle listings to uncover actionable insights that support strategic pricing, targeted marketing, and competitive positioning in the used car market.

Data Description

Aggregated dataset complied from multiple online automotive marketplace listings.

Observations: 98,837

Variables (10):

• Categorical: Brand, Model, Transmission, Fuel Type

• Numeric: Year, Price, Mileage, Tax, MPG, Engine Size

Summary Statistics:

• Average price: $16,797

• Average mileage: 23,094 miles

• Most common fuel type: Petrol

• Transmission distribution: 57% manual, 43% automatic

Data Preprocessing & Exploratory Data Analysis(EDA)

# Importing Libraries

!pip install statsmodels
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import plotly.express as px
import polars as pl

from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.cluster import KMeans
from sklearn.linear_model import LinearRegression
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_squared_error, r2_score
from sklearn.metrics import classification_report, confusion_matrix, accuracy_score
from sklearn.decomposition import PCA
import joblib
from sklearn.pipeline import Pipeline
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.ensemble import RandomForestRegressor
from sklearn.svm import SVR
from sklearn.linear_model import LogisticRegression

Defaulting to user installation because normal site-packages is not writeable
Requirement already satisfied: statsmodels in /home/vscode/.local/lib/python3.12/site-packages (0.14.6)
Requirement already satisfied: numpy<3,>=1.22.3 in /home/vscode/.local/lib/python3.12/site-packages (from statsmodels) (2.3.2)
Requirement already satisfied: scipy!=1.9.2,>=1.8 in /home/vscode/.local/lib/python3.12/site-packages (from statsmodels) (1.16.1)
Requirement already satisfied: pandas!=2.1.0,>=1.4 in /home/vscode/.local/lib/python3.12/site-packages (from statsmodels) (2.3.2)
Requirement already satisfied: patsy>=0.5.6 in /home/vscode/.local/lib/python3.12/site-packages (from statsmodels) (1.0.2)
Requirement already satisfied: packaging>=21.3 in /home/vscode/.local/lib/python3.12/site-packages (from statsmodels) (25.0)
Requirement already satisfied: python-dateutil>=2.8.2 in /home/vscode/.local/lib/python3.12/site-packages (from pandas!=2.1.0,>=1.4->statsmodels) (2.9.0.post0)
Requirement already satisfied: pytz>=2020.1 in /home/vscode/.local/lib/python3.12/site-packages (from pandas!=2.1.0,>=1.4->statsmodels) (2025.2)
Requirement already satisfied: tzdata>=2022.7 in /home/vscode/.local/lib/python3.12/site-packages (from pandas!=2.1.0,>=1.4->statsmodels) (2025.2)
Requirement already satisfied: six>=1.5 in /home/vscode/.local/lib/python3.12/site-packages (from python-dateutil>=2.8.2->pandas!=2.1.0,>=1.4->statsmodels) (1.17.0)

# Loading and Cleaning Data with Polars

# Load data
cars = pl.read_csv("Final_dataset.csv", ignore_errors= True)

cars = cars.drop_nulls()

# View the first few rows of the data
cars.head() 

# Summary of dataset
# Reason: Quick overview of the cleaned dataset.

print("Shape:", cars.shape)     # number of rows and columns
print("\nColumn types:")
print(cars.schema)              # check data types

print("\nSummary statistics:")

cars.describe()

cars.head() 

Shape: (98836, 10)

Column types:
Schema({'Brand': String, 'model': String, 'year': Int64, 'price': Int64, 'transmission': String, 'mileage': Int64, 'fuelType': String, 'tax': Int64, 'mpg': Float64, 'engineSize': Float64})

Summary statistics:

shape: (5, 10)

Brand	model	year	price	transmission	mileage	fuelType	tax	mpg	engineSize
str	str	i64	i64	str	i64	str	i64	f64	f64
"VW"	"T-Roc"	2019	25000	"Automatic"	13904	"Diesel"	145	49.6	2.0
"VW"	"T-Roc"	2019	26883	"Automatic"	4562	"Diesel"	145	49.6	2.0
"VW"	"T-Roc"	2019	20000	"Manual"	7414	"Diesel"	145	50.4	2.0
"VW"	"T-Roc"	2019	33492	"Automatic"	4825	"Petrol"	145	32.5	2.0
"VW"	"T-Roc"	2019	22900	"Semi-Auto"	6500	"Petrol"	150	39.8	1.5

Data Description Dashboard

from plotly.subplots import make_subplots
import plotly.graph_objects as go

# Calculate key metrics
total_vehicles = cars.shape[0]
avg_price = cars["price"].mean()
median_price = cars["price"].median()
avg_mileage = cars["mileage"].mean()
avg_mpg = cars["mpg"].mean()
num_brands = cars["Brand"].n_unique()
avg_age = 2025 - cars["year"].mean()

# Prepare data for charts
brand_counts = cars.group_by("Brand").agg(pl.len().alias("count")).sort("count", descending=True).head(6)
fuel_counts = cars.group_by("fuelType").agg(pl.len().alias("count")).sort("count", descending=True)
trans_counts = cars.group_by("transmission").agg(pl.len().alias("count"))

# Define color palette
colors = {
    "primary": "#4361EE",
    "secondary": "#3A0CA3", 
    "accent1": "#7209B7",
    "accent2": "#F72585",
    "accent3": "#4CC9F0",
    "success": "#06D6A0",
    "warning": "#FFD166",
    "dark": "#1A1A2E",
    "light": "#F8F9FA",
    "gradient": ["#4361EE", "#3A0CA3", "#7209B7", "#F72585", "#4CC9F0", "#06D6A0"]
}

# Create subplot layout
fig = make_subplots(
    rows=3, cols=3,
    specs=[
        [{"type": "indicator"}, {"type": "indicator"}, {"type": "indicator"}],
        [{"type": "indicator"}, {"type": "indicator"}, {"type": "indicator"}],
        [{"type": "bar"}, {"type": "pie"}, {"type": "pie"}]
    ],
    subplot_titles=(
        "", "", "",
        "", "", "",
        "<b>Top Brands by Inventory</b>", "<b>Fuel Type Mix</b>", "<b>Transmission Split</b>"
    ),
    vertical_spacing=0.10,
    horizontal_spacing=0.06,
    row_heights=[0.28, 0.28, 0.44]
)

# Row 1: KPI Indicators
fig.add_trace(go.Indicator(
    mode="number",
    value=total_vehicles,
    title={
        "text": "<b style='font-size:16px;color:#1A1A2E'>Total Vehicles</b><br><span style='font-size:11px;color:#6c757d'>Listed in Market</span>",
        "font": {"size": 14}
    },
    number={"font": {"size": 42, "color": colors["primary"], "family": "Arial Black"}, "valueformat": ","},
    domain={"x": [0, 1], "y": [0, 1]}
), row=1, col=1)

fig.add_trace(go.Indicator(
    mode="number",
    value=avg_price,
    title={
        "text": "<b style='font-size:16px;color:#1A1A2E'>Average Price</b><br><span style='font-size:11px;color:#6c757d'>Market Mean</span>",
        "font": {"size": 14}
    },
    number={"font": {"size": 42, "color": colors["success"], "family": "Arial Black"}, "prefix": "$", "valueformat": ",.0f"}
), row=1, col=2)

fig.add_trace(go.Indicator(
    mode="number",
    value=median_price,
    title={
        "text": "<b style='font-size:16px;color:#1A1A2E'>Median Price</b><br><span style='font-size:11px;color:#6c757d'>Typical Vehicle</span>",
        "font": {"size": 14}
    },
    number={"font": {"size": 42, "color": colors["accent3"], "family": "Arial Black"}, "prefix": "$", "valueformat": ",.0f"}
), row=1, col=3)

# Row 2: More KPIs
fig.add_trace(go.Indicator(
    mode="number",
    value=avg_mileage,
    title={
        "text": "<b style='font-size:16px;color:#1A1A2E'>Avg Mileage</b><br><span style='font-size:11px;color:#6c757d'>Miles Driven</span>",
        "font": {"size": 14}
    },
    number={"font": {"size": 42, "color": colors["warning"], "family": "Arial Black"}, "valueformat": ",.0f"}
), row=2, col=1)

fig.add_trace(go.Indicator(
    mode="number",
    value=num_brands,
    title={
        "text": "<b style='font-size:16px;color:#1A1A2E'>Brands</b><br><span style='font-size:11px;color:#6c757d'>In Market</span>",
        "font": {"size": 14}
    },
    number={"font": {"size": 42, "color": colors["accent1"], "family": "Arial Black"}}
), row=2, col=2)

fig.add_trace(go.Indicator(
    mode="number",
    value=round(avg_age, 1),
    title={
        "text": "<b style='font-size:16px;color:#1A1A2E'>Avg Vehicle Age</b><br><span style='font-size:11px;color:#6c757d'>Years Old</span>",
        "font": {"size": 14}
    },
    number={"font": {"size": 42, "color": colors["accent2"], "family": "Arial Black"}, "suffix": " yrs"}
), row=2, col=3)

# Row 3: Charts
# Top Brands Bar Chart - Horizontal with gradient
brand_list = brand_counts["Brand"].to_list()
count_list = brand_counts["count"].to_list()

fig.add_trace(go.Bar(
    y=brand_list[::-1],
    x=count_list[::-1],
    orientation="h",
    marker=dict(
        color=count_list[::-1],
        colorscale=[[0, "#4CC9F0"], [0.5, "#4361EE"], [1, "#3A0CA3"]],
        line=dict(width=0),
        cornerradius=5
    ),
    text=[f"{x:,}" for x in count_list[::-1]],
    textposition="outside",
    textfont=dict(size=11, color="#1A1A2E", family="Arial")
), row=3, col=1)

# Fuel Type Donut - Enhanced
fig.add_trace(go.Pie(
    labels=fuel_counts["fuelType"].to_list(),
    values=fuel_counts["count"].to_list(),
    hole=0.55,
    marker=dict(
        colors=["#4361EE", "#06D6A0", "#FFD166", "#F72585", "#4CC9F0"],
        line=dict(color="#FFFFFF", width=2)
    ),
    textinfo="percent",
    textfont=dict(size=12, color="white", family="Arial Black"),
    hovertemplate="<b>%{label}</b><br>Count: %{value:,}<br>Share: %{percent}<extra></extra>",
    rotation=90
), row=3, col=2)

# Transmission Donut - Enhanced
fig.add_trace(go.Pie(
    labels=trans_counts["transmission"].to_list(),
    values=trans_counts["count"].to_list(),
    hole=0.55,
    marker=dict(
        colors=["#7209B7", "#4CC9F0"],
        line=dict(color="#FFFFFF", width=2)
    ),
    textinfo="percent+label",
    textfont=dict(size=12, color="white", family="Arial"),
    hovertemplate="<b>%{label}</b><br>Count: %{value:,}<br>Share: %{percent}<extra></extra>",
    rotation=45
), row=3, col=3)

# Update layout
fig.update_layout(
    height=800,
    width=1200,
    title=dict(
        text="<b>🚗 USED CAR MARKET DASHBOARD</b><br><span style='font-size:14px;color:#6c757d;font-weight:normal'>Strategic Market Overview & Key Performance Indicators</span>",
        font=dict(size=26, color="#1A1A2E", family="Arial Black"),
        x=0.5,
        xanchor="center",
        y=0.97
    ),
    showlegend=False,
    margin=dict(l=60, r=60, t=120, b=40),
    paper_bgcolor="#FFFFFF",
    plot_bgcolor="#FFFFFF",
    font=dict(family="Arial", color="#1A1A2E")
)

# Style subplot titles
for annotation in fig['layout']['annotations']:
    annotation['font'] = dict(size=13, color="#1A1A2E", family="Arial")

# Update axes for bar chart
fig.update_xaxes(
    showgrid=True,
    gridcolor="#E9ECEF",
    gridwidth=1,
    zeroline=False,
    tickfont=dict(size=10),
    title_text="",
    row=3, col=1
)
fig.update_yaxes(
    showgrid=False,
    tickfont=dict(size=11, color="#1A1A2E"),
    title_text="",
    row=3, col=1
)

# Add subtle border effect with shapes
fig.add_shape(
    type="rect",
    xref="paper", yref="paper",
    x0=0, y0=0, x1=1, y1=1,
    line=dict(color="#E9ECEF", width=2)
)

# Add section divider line
fig.add_shape(
    type="line",
    xref="paper", yref="paper",
    x0=0.02, y0=0.42, x1=0.98, y1=0.42,
    line=dict(color="#E9ECEF", width=1, dash="dot")
)

fig.show()

Visualization: Average Price by Brand (EDA)

# Summary stats by brand
brand_summary = (
    cars.group_by("Brand")
        .agg([
            pl.mean("price").alias("avg_price"),
            pl.mean("mileage").alias("avg_mileage"),
            pl.mean("engineSize").alias("avg_engineSize"),
            pl.mean("mpg").alias("avg_mpg"),
            pl.len().alias("count")
        ])
        .sort("avg_price", descending=True)
)

brand_summary

fig = px.bar(
    brand_summary.to_pandas(),
    x="Brand",
    y="avg_price",
    title="Average Price by Brand",
    labels={"avg_price": "Average Price ($)"},
)
fig.show()

Price Distribution in Used Car Market (EDA)

# Filter out extreme outliers for better visualization
cars_filtered = cars.filter(pl.col("price") <= 50000)

fig = px.histogram(
    cars_filtered.to_pandas(),
    x="price",
    nbins=50,
    title="Price Distribution in Used Car Market",
    labels={"price": "Price ($)", "count": "Number of Vehicles"},
    color_discrete_sequence=["#3498db"]
)

fig.add_vline(
    x=cars_filtered["price"].mean(), 
    line_dash="dash", 
    line_color="red",
    annotation_text=f"Mean: ${cars_filtered['price'].mean():,.0f}",
    annotation_position="top right"
)

fig.add_vline(
    x=cars_filtered["price"].median(), 
    line_dash="dot", 
    line_color="green",
    annotation_text=f"Median: ${cars_filtered['price'].median():,.0f}",
    annotation_position="top left"
)

fig.update_layout(
    height=500,
    width=1000,
    title=dict(
        text="<b>Price Distribution in Used Car Market</b>",
        font=dict(size=20),
        x=0.5,
        xanchor="center"
    ),
    xaxis=dict(
        title="Price ($)",
        tickprefix="$",
        tickformat=",.0f"
    ),
    yaxis=dict(title="Number of Vehicles"),
    margin=dict(l=80, r=50, t=80, b=60)
)

fig.show()

# Note how many cars were filtered
total = cars.shape[0]
filtered = cars_filtered.shape[0]
print(f"Showing {filtered:,} of {total:,} vehicles ({filtered/total*100:.1f}%) - filtered to prices ≤ $50,000")

Showing 97,686 of 98,836 vehicles (98.8%) - filtered to prices ≤ $50,000

Key Findings

Right-Skewed Distribution:

The price distribution exhibits a classic right-skewed pattern, with most vehicles clustered in lower price ranges and a long tail extending toward premium prices. This reflects a market dominated by affordable, mass-market vehicles.

Mean vs Median Gap:

The mean price exceeds the median, confirming the presence of high-priced outliers likely luxury and premium vehicles that inflate the average. The median serves as a more accurate measure of typical used car pricing.

Price Concentration:

The majority of inventory falls within the $10,000–$25,000 range, representing the market’s sweet spot where buyers seek the best balance of value and reliability.

Market Accessibility:

A substantial portion of vehicles priced under $15,000 indicates strong availability for budget-conscious consumers and first-time buyers entering the market.

Premium Segment:

Vehicles exceeding $35,000 represent a smaller but significant segment, typically comprising newer models, luxury brands, and low-mileage vehicles.

Marketing Implications:

Pricing Strategy:

Dealerships should concentrate inventory acquisition and marketing efforts on the $10K–$25K range where buyer demand is strongest.

Segmented campaigns:

Create unique marketing strategies: value-focused messaging for budget consumers (less $15K) and premium positioning for luxury purchasers (above $35K).

Competitive Positioning:

Vehicles priced near the median offer optimal balance between market competitiveness and profit margin potential.

Financing Options:

Promote financing solutions for vehicles above the median price to increase accessibility and expand the potential buyer pool.

Visualization: Market Share By Brand (EDA)

fig = px.pie(
    brand_summary.to_pandas(),
    names="Brand",
    values="count",
    title="Market Share by Brand (Count of Listings)"
)
fig.show()

Market Segmentation using Clustering (k-means analysis)

# Select features for clustering
cluster_features = ['year', 'price', 'mileage', 'tax', 'mpg', 'engineSize']
cars_cluster = cars.select(cluster_features).drop_nulls()

# Create clustering pipeline
def create_kmeans_pipeline(n_clusters, random_state=42):
    return Pipeline([
        ('scaler', StandardScaler()),
        ('kmeans', KMeans(n_clusters=n_clusters, random_state=random_state, n_init=10))
    ])

We use Customer segmentation model to investigate the data set.

Select relevant features for clustering

cars_bases = cars.select([
 #'Brand',
 #'model',
 'year',
 'price',
 #'transmission',
 'mileage',
 #'fuelType',
 'tax',
 'mpg',
 'engineSize'])

Creating K Means clustering pipeline

def create_pipeline(num_clusters, random_seed = 42):
    """
    Creates a machine learning pipeline with a scaler and KMeans.
    """
    pipeline = Pipeline([
        ('scaler', StandardScaler()),
        ('kmeans', KMeans(n_clusters=num_clusters, random_state=random_seed))
    ])
    return pipeline  

Determining Optimal clusters through Elbow Plot

def calculate_totwithinss(data, k):
    kmeans_pipeline = create_pipeline(k, random_seed=10)
    kmeans_pipeline.fit(data)
    return kmeans_pipeline['kmeans'].inertia_

# Calculate tot.withinss for different values of k
k_values = range(1, 10)
totwithinss_values = [calculate_totwithinss(cars_bases, k) for k in k_values]

# Create a DataFrame for results
kmeans_results = pl.DataFrame(
    {'num_clusters': k_values,
     'tot_withinss': totwithinss_values})

# Plot the elbow method using Plotly Express
elbow_plot = px.line(
    data_frame = kmeans_results,
    x = 'num_clusters',
    y = 'tot_withinss', 
    markers = True,
    labels = {
        'num_clusters': 'Number of Clusters', 'tot_withinss': 'Total Within SS'
        },
    title = 'Elbow Method for Optimal k')

elbow_plot.show()

Based on the elbow method, four clusters were chosen because they provide the best balance between model simplicity and meaningful separation of distinct vehicle groups in the used car market.

K Means Clustering

# Choose the number of clusters based on the elbow method
optimal_k = 4

# Run K-means clustering

cars_kmeans_pipeline = create_pipeline(optimal_k)
cars_kmeans_pipeline.fit(cars_bases)


# Add cluster assignments to the original data
cars_with_clusters = cars.with_columns(
    pl.Series(
        "segment_number",
        cars_kmeans_pipeline['kmeans'].labels_ + 1
        ).cast(pl.Utf8).cast(pl.Categorical)  # Make cluster labels 1-indexed
)

cars_with_clusters.head()

shape: (5, 11)

Brand	model	year	price	transmission	mileage	fuelType	tax	mpg	engineSize	segment_number
str	str	i64	i64	str	i64	str	i64	f64	f64	cat
"VW"	"T-Roc"	2019	25000	"Automatic"	13904	"Diesel"	145	49.6	2.0	"3"
"VW"	"T-Roc"	2019	26883	"Automatic"	4562	"Diesel"	145	49.6	2.0	"3"
"VW"	"T-Roc"	2019	20000	"Manual"	7414	"Diesel"	145	50.4	2.0	"1"
"VW"	"T-Roc"	2019	33492	"Automatic"	4825	"Petrol"	145	32.5	2.0	"3"
"VW"	"T-Roc"	2019	22900	"Semi-Auto"	6500	"Petrol"	150	39.8	1.5	"1"

Segment Description:

Analyzing the segments based on mean values of key metrics and the number of observations in each segment.

# Calculating summary statistics for each segment

segment_summary = cars_with_clusters.group_by('segment_number').agg(
    [
        pl.mean('price').alias('mean_price'),
        pl.mean('mileage').alias('mean_mileage'),
        pl.mean('engineSize').alias('mean_engineSize'),
        pl.len().alias('n')
    ]
)

segment_summary

shape: (4, 5)

segment_number	mean_price	mean_mileage	mean_engineSize	n
cat	f64	f64	f64	u32
"2"	11232.150296	57132.047022	1.957311	10293
"1"	14764.997357	15145.826807	1.415791	47294
"3"	32028.494608	9038.880616	2.308308	18453
"4"	11197.826241	35593.776145	1.522789	22796

Customer Segmentation Conclusion

The customer segmentation analysis successfully identified four clusters,through the elbow method.

Cluster 1:

Represents mid-range cars with moderate prices (~$14,765), average mileage (~15,146), and smaller engines (~1.42 L). Likely balanced options between affordability and performance.

Cluster 2:

Consists of budget or older vehicles with the lowest mean price (~$11,232) but highest mileage (~57,132), indicating heavy usage and smaller engine capacity (~1.96 L).

Cluster 3:

Represents premium or high-performance vehicles, having the highest mean price (~$32,028), lowest mileage (~9,039), and largest engine size (~2.31 L) — suggesting newer or luxury cars.

Cluster 4:

Also economy-oriented, with a similar low price (~$11,198) and moderate mileage (~35,594). Slightly smaller engines (~1.52 L), possibly compact cars.

PCA Visualization of K-Means Clustering

from sklearn.decomposition import PCA

# Standardize the features manually for PCA
scaler = StandardScaler()
scaled_features = scaler.fit_transform(cars_bases.to_pandas())

# Reduce to 2 dimensions
pca = PCA(n_components=2)
pca_components = pca.fit_transform(scaled_features)

# Add PCA components + cluster labels to dataframe
pca_df = pd.DataFrame({
    "PC1": pca_components[:, 0],
    "PC2": pca_components[:, 1],
    "Cluster": cars_kmeans_pipeline['kmeans'].labels_ + 1
})

Plot PCA Visualization

fig = px.scatter(
    pca_df,
    x="PC1", 
    y="PC2",
    color="Cluster",
    title="PCA Visualization of K-Means Clusters",
    opacity=0.8
)

fig.show()

Interpretation :

PCA is a dimensionality reduction technique. When we have many variables (price, mileage, engine size, mpg, year, tax), it becomes difficult to plot them all together. Since clustering happens in a 6-dimensional space, PCA reduces it to a 2-D plot so humans can visualize the groupings.

1. Axes meaning (PC1 and PC2)

PC1 (horizontal axis) captures the largest pattern in the data. It often represents a combined effect of price + engine size + year (newer, premium cars on one side, older budget cars on the other).

PC2 (vertical axis) captures the second largest pattern. Often relates to mileage + efficiency variation.

Visualizations for Market Segmentation (dashboard)

STEP 1: PREPARE DATA FOR CLUSTERING

# Convert Polars to Pandas (add this before your clustering code)
df = cars.to_pandas()

# Then your clustering code will work
cluster_features = ['price', 'mileage', 'mpg', 'engineSize', 'year']
X = df[cluster_features].dropna()

print(f"\nClustering on {len(X):,} vehicles with features: {cluster_features}")

scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

Clustering on 98,836 vehicles with features: ['price', 'mileage', 'mpg', 'engineSize', 'year']

STEP 2: PERFORM K-MEANS CLUSTERING

print("\n" + "="*60)
print("STEP 2: K-MEANS CLUSTERING (K=4)")
print("="*60)

optimal_k = 4
kmeans = KMeans(n_clusters=optimal_k, random_state=42, n_init=10)
clusters = kmeans.fit_predict(X_scaled)

df_clustered = df.loc[X.index].copy()
df_clustered['Cluster'] = clusters

print(f"Clustering complete! {optimal_k} segments identified.")

============================================================
STEP 2: K-MEANS CLUSTERING (K=4)
============================================================
Clustering complete! 4 segments identified.

STEP 3: ANALYZE CLUSTER PROFILES

print("\n" + "="*60)
print("STEP 3: CLUSTER PROFILES")
print("="*60)

cluster_stats = df_clustered.groupby('Cluster').agg({
    'price': ['mean', 'median', 'min', 'max', 'count'],
    'mileage': 'mean',
    'mpg': 'mean',
    'engineSize': 'mean',
    'year': 'mean'
}).round(2)

cluster_stats.columns = ['Avg Price', 'Median Price', 'Min Price', 'Max Price', 'Count', 
                          'Avg Mileage', 'Avg MPG', 'Avg Engine Size', 'Avg Car Age']

cluster_stats = cluster_stats.sort_values('Avg Price')

segment_names = ['Budget', 'Economy', 'Mid-Range', 'Premium']
cluster_stats['Segment'] = segment_names

cluster_to_segment = dict(zip(cluster_stats.index, segment_names))
df_clustered['Segment'] = df_clustered['Cluster'].map(cluster_to_segment)

print("\nCluster Profiles:")
print(cluster_stats.to_string())

============================================================
STEP 3: CLUSTER PROFILES
============================================================

Cluster Profiles:
         Avg Price  Median Price  Min Price  Max Price  Count  Avg Mileage  Avg MPG  Avg Engine Size  Avg Car Age    Segment
Cluster                                                                                                                     
0         10727.37       10000.0        450      37999  17009     57829.50    57.89             1.85      2014.08     Budget
2         11952.43       11295.0       3334      52495  44093     21472.02    61.09             1.37      2016.93    Economy
1         22445.48       21980.0       6495      67940  31972      8462.09    48.25             1.72      2018.72  Mid-Range
3         40452.77       39485.0      14995     159999   5762     14165.08    39.10             3.06      2018.03    Premium

STEP 4: VISUALIZATIONS

print("\n" + "="*60)
print("STEP 4: VISUALIZATIONS")
print("="*60)

colors = {'Budget': '#27ae60', 'Economy': '#3498db', 'Mid-Range': '#f39c12', 'Premium': '#e74c3c'}
segment_order = ['Budget', 'Economy', 'Mid-Range', 'Premium']

fig, axes = plt.subplots(1, 3, figsize=(18, 6))


# Plot 1: Average Price by Segment (Bar Chart)
ax1 = axes[0]
segment_avg = df_clustered.groupby('Segment')['price'].mean().reindex(segment_order)

bars = ax1.bar(segment_order, segment_avg, color=[colors[s] for s in segment_order], edgecolor='black')
ax1.set_xlabel('Segment', fontsize=12)
ax1.set_ylabel('Average Price ($)', fontsize=12)
ax1.set_title('Average Price by Market Segment', fontsize=14, fontweight='bold')

for bar, val in zip(bars, segment_avg):
    ax1.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 500, 
             f'${val:,.0f}', ha='center', fontsize=11, fontweight='bold')

ax1.grid(True, alpha=0.3, axis='y')


# Plot 2: Price vs Mileage Scatter
ax2 = axes[1]
for segment in segment_order:
    seg_data = df_clustered[df_clustered['Segment'] == segment]
    ax2.scatter(seg_data['mileage'], seg_data['price'], 
                c=colors[segment], label=segment, alpha=0.5, s=20, edgecolors='white', linewidth=0.5)

ax2.set_xlabel('Mileage', fontsize=12)
ax2.set_ylabel('Price ($)', fontsize=12)
ax2.set_title('Market Segments: Price vs Mileage', fontsize=14, fontweight='bold')
ax2.legend(title='Segment', fontsize=10)
ax2.grid(True, alpha=0.3)


# Plot 3: Segment Distribution (Pie Chart)
ax3 = axes[2]
segment_counts = df_clustered['Segment'].value_counts().reindex(segment_order)
explode = (0.02, 0.02, 0.02, 0.05)

wedges, texts, autotexts = ax3.pie(segment_counts, labels=segment_order, autopct='%1.1f%%',
                                    colors=[colors[s] for s in segment_order], explode=explode,
                                    startangle=90, textprops={'fontsize': 11})
ax3.set_title('Market Segment Distribution', fontsize=14, fontweight='bold')

plt.tight_layout()
plt.savefig('market_segmentation.png', dpi=300, bbox_inches='tight')
plt.show()

print("Segmentation visualization saved as 'market_segmentation.png'")

============================================================
STEP 4: VISUALIZATIONS
============================================================

Segmentation visualization saved as 'market_segmentation.png'

1.Price Distribution Across Clusters

Cluster 3:

Shows the highest prices and the largest spread, with many extreme high-value outliers. This segment represents premium buyers willing to spend significantly more. Ideal for luxury products, premium packages, and high-value upselling.

Cluster 1:

Falls into the mid-range with stable price behavior. These customers look for value but are not price-sensitive, making them good targets for balanced offers, loyalty programs, and add-on bundles.

Cluster 4:

Shows lower-mid pricing, with a relatively tight distribution. These customers prefer affordable but decent-quality options. Best suited for discount campaigns, value packs, and entry-level products.

Cluster 2:

Has the lowest median prices and limited high-price purchases. This segment is the most budget-conscious, responding well to price cuts, seasonal promotions, and low-cost alternatives.

2.Mileage Distribution Across Clusters

Cluster 2:

Has the highest mileage and the widest spread, indicating older or heavily used vehicles. This segment is ideal for maintenance plans, repair services, and high-wear part replacements.

Cluster 4:

Shows moderate-to-high mileage, suggesting frequent drivers. They are strong candidates for routine service reminders, safety checks, and periodic maintenance offers.

Cluster 1:

Displays mid-range mileage, reflecting balanced usage. They fit well with standard service intervals, tune-up promotions, and value-based maintenance packages.

Cluster 3:

Has the lowest mileage, indicating newer or lightly used vehicles. Marketing should emphasize premium upgrades, accessories, detailing services, and extended warranties, not heavy repair-oriented messaging.

3.Engine Size Distribution Across Clusters

Cluster 3:

Shows the largest engine sizes, including many high-end outliers. This indicates customers who prefer powerful, performance-oriented vehicles. Ideal for marketing premium performance packages, fuel additives, and high-end service plans.

Cluster 1:

Has the smallest and most compact engine sizes, with a tight distribution around lower engine ranges. These customers favor economical, fuel-efficient vehicles, making them good targets for budget maintenance services and efficiency-focused offers.

Cluster 4:

Shows moderate engine sizes, indicating a balance between power and efficiency. They represent a versatile group suited for general service promotions, mid-range upgrades, and value-focused maintenance.

Cluster 2:

Has small-to-mid engine sizes with some higher outliers. This group mixes practicality with occasional preference for slightly more powerful vehicles. Marketing can focus on balanced service packages and flexible upgrade options.

Targeting

car = cars_with_clusters.to_pandas()

X = car[['year','price','mileage','tax','mpg','engineSize']]
y = car['segment_number'].astype(int)

Bar Plot for Feature Importance

# Train/Test Split
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42, stratify=y
)

Build Model Pipeline

clf_pipeline = Pipeline([
    ('scaler', StandardScaler()),
    ('model', LogisticRegression(max_iter=2000, multi_class='auto'))
])

Fit model

clf_pipeline.fit(X_train, y_train)

/home/vscode/.local/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:1272: FutureWarning:

'multi_class' was deprecated in version 1.5 and will be removed in 1.7. From then on, it will always use 'multinomial'. Leave it to its default value to avoid this warning.

Pipeline(steps=[('scaler', StandardScaler()),
                ('model',
                 LogisticRegression(max_iter=2000, multi_class='auto'))])

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

BRAND POSITIONING (Binary Outcome Prediction) Premium vs Non-Premium Cars

# Create binary target: Premium vs Non-Premium
car_bin = cars_with_clusters.to_pandas()
car_bin['premium_flag'] = (car_bin['segment_number'] == '3').astype(int)

# Define features
X = car_bin[['year','price','mileage','tax','mpg','engineSize']]
y = car_bin['premium_flag']

# Split data
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42, stratify=y)

# Model
from sklearn.linear_model import LogisticRegression
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler

binary_pipe = Pipeline([
    ('scaler', StandardScaler()),
    ('model', LogisticRegression())
])

binary_pipe.fit(X_train, y_train)

Pipeline(steps=[('scaler', StandardScaler()), ('model', LogisticRegression())])

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

Predictions

y_pred = binary_pipe.predict(X_test)

from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

print("Accuracy:", accuracy_score(y_test, y_pred))
print("Classification Report:\n", classification_report(y_test, y_pred))
confusion_matrix(y_test, y_pred)

Accuracy: 0.9790900812788776
Classification Report:
               precision    recall  f1-score   support

           0       0.99      0.99      0.99     24115
           1       0.95      0.94      0.94      5536

    accuracy                           0.98     29651
   macro avg       0.97      0.96      0.97     29651
weighted avg       0.98      0.98      0.98     29651

array([[23833,   282],
       [  338,  5198]])

The logistic regression model accurately predicts premium vs. non-premium cars (98% accuracy), with very low misclassification rates, making it highly reliable for pricing, segmentation, and inventory strategy

import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)

plt.figure(figsize=(6,4))
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues',
            xticklabels=['Non-Premium','Premium'],
            yticklabels=['Non-Premium','Premium'])
plt.title("Confusion Matrix for Premium Car Classification")
plt.xlabel("Predicted")
plt.ylabel("Actual")
plt.show()

import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix

importance = binary_pipe.named_steps['model'].coef_[0]
features = X.columns

imp_df = pd.DataFrame({
    "feature": features,
    "importance": importance
}).sort_values(by="importance", ascending=False)

plt.figure(figsize=(7,4))
sns.barplot(data=imp_df, x="importance", y="feature", palette="viridis")
plt.title("Feature Importance for Premium Prediction (Logistic Regression)")
plt.xlabel("Coefficient Value")
plt.ylabel("Feature")
plt.show()

/tmp/ipykernel_3569/1751013292.py:14: FutureWarning:



Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect.

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.metrics import accuracy_score, roc_auc_score, classification_report
import warnings
warnings.filterwarnings('ignore')

# Set professional style
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['font.sans-serif'] = ['Arial']

print("=" * 80)
print("LOGISTIC REGRESSION - FEATURE IMPORTANCE ANALYSIS")
print("=" * 80)

# Convert to pandas and clean
df = cars.to_pandas()
df = df.dropna(subset=['mileage', 'year', 'engineSize', 'mpg', 'tax', 
                       'transmission', 'fuelType', 'price'])

print(f"\nDataset: {len(df):,} vehicles (after removing missing values)")

# ============================================================================
# CREATE PREMIUM CLASSIFICATION TARGET
# ============================================================================
premium_threshold = df['price'].quantile(0.70)
df['is_premium'] = (df['price'] >= premium_threshold).astype(int)

print(f"\nPremium Threshold: ${premium_threshold:,.0f}")
print(f"Premium vehicles: {df['is_premium'].sum():,} ({df['is_premium'].mean()*100:.1f}%)")
print(f"Non-Premium vehicles: {(~df['is_premium'].astype(bool)).sum():,} ({(1-df['is_premium'].mean())*100:.1f}%)")

# ============================================================================
# ENCODE CATEGORICAL VARIABLES
# ============================================================================
le_trans = LabelEncoder()
le_fuel = LabelEncoder()

df['transmission_encoded'] = le_trans.fit_transform(df['transmission'])
df['fuelType_encoded'] = le_fuel.fit_transform(df['fuelType'])

print("\nTransmission encoding:")
for i, trans in enumerate(le_trans.classes_):
    print(f"  {trans}: {i}")

print("\nFuel type encoding:")
for i, fuel in enumerate(le_fuel.classes_):
    print(f"  {fuel}: {i}")

# ============================================================================
# PREPARE FEATURES (CORE FEATURES ONLY)
# ============================================================================
feature_columns = ['mileage', 'year', 'engineSize', 'mpg', 'tax', 
                   'transmission_encoded', 'fuelType_encoded']

X = df[feature_columns]
y = df['is_premium']

print(f"\nFeatures: {len(feature_columns)}")
print(f"Samples: {len(X):,}")

# ============================================================================
# TRAIN-TEST SPLIT
# ============================================================================
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42, stratify=y
)

print(f"\nTraining set: {len(X_train):,} samples")
print(f"Test set: {len(X_test):,} samples")

# ============================================================================
# FEATURE SCALING (REQUIRED FOR LOGISTIC REGRESSION)
# ============================================================================
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)

print("\n✓ Features standardized")

# ============================================================================
# TRAIN LOGISTIC REGRESSION MODEL
# ============================================================================
print("\n" + "-" * 80)
print("TRAINING LOGISTIC REGRESSION MODEL")
print("-" * 80)

# Train model
log_reg = LogisticRegression(random_state=42, max_iter=2000, C=1.0, solver='lbfgs')
log_reg.fit(X_train_scaled, y_train)

# Make predictions
y_pred = log_reg.predict(X_test_scaled)
y_proba = log_reg.predict_proba(X_test_scaled)[:, 1]

# Calculate metrics
accuracy = accuracy_score(y_test, y_pred)
roc_auc = roc_auc_score(y_test, y_proba)

print(f"\n✓ Model trained successfully")
print(f"\nPerformance Metrics:")
print(f"  Accuracy: {accuracy:.4f} ({accuracy*100:.2f}%)")
print(f"  ROC AUC: {roc_auc:.4f}")

# ============================================================================
# EXTRACT FEATURE IMPORTANCE (COEFFICIENTS)
# ============================================================================
coefficients = log_reg.coef_[0]

feature_importance = pd.DataFrame({
    'feature': feature_columns,
    'coefficient': coefficients,
    'abs_coefficient': np.abs(coefficients)
}).sort_values('abs_coefficient', ascending=False)

print(f"\n" + "-" * 80)
print("FEATURE IMPORTANCE (LOGISTIC REGRESSION COEFFICIENTS)")
print("-" * 80)
print(f"\n{'Rank':<6} {'Feature':<25} {'Coefficient':<15} {'Interpretation':<20}")
print("-" * 80)

for idx, (i, row) in enumerate(feature_importance.iterrows()):
    direction = "Positive ↑" if row['coefficient'] > 0 else "Negative ↓"
    print(f"{idx+1:<6} {row['feature']:<25} {row['coefficient']:>+8.4f}      {direction}")

print("-" * 80)

# ============================================================================
# CREATE VISUALIZATION
# ============================================================================
fig, ax = plt.subplots(figsize=(14, 8))

# Sort by coefficient value (not absolute) for better visual
plot_data = feature_importance.sort_values('coefficient', ascending=True)
values = plot_data['coefficient'].values

# Create readable feature names
feature_names = []
for feat in plot_data['feature']:
    if feat == 'transmission_encoded':
        feature_names.append('Transmission Type')
    elif feat == 'fuelType_encoded':
        feature_names.append('Fuel Type')
    elif feat == 'engineSize':
        feature_names.append('Engine Size')
    elif feat == 'year':
        feature_names.append('Year')
    elif feat == 'mileage':
        feature_names.append('Mileage')
    elif feat == 'mpg':
        feature_names.append('MPG')
    elif feat == 'tax':
        feature_names.append('Tax')
    else:
        feature_names.append(feat.capitalize())

# Color by positive/negative impact
colors_feat = ['#2A9D8F' if v > 0 else '#E76F51' for v in values]

# Create horizontal bar chart
bars = ax.barh(range(len(values)), values, 
              color=colors_feat, alpha=0.85, edgecolor='black', linewidth=2.5)

# Customize axes
ax.set_yticks(range(len(values)))
ax.set_yticklabels(feature_names, fontsize=13, fontweight='bold')
ax.set_xlabel('Coefficient Value', fontsize=14, fontweight='bold', color='#0d1b2a')
ax.set_title('Feature Importance for Premium Prediction\n(Logistic Regression Model)', 
            fontsize=16, fontweight='bold', pad=20, color='#0d1b2a')

# Add zero line
ax.axvline(x=0, color='black', linewidth=2.5, linestyle='--', alpha=0.7)

# Grid and background
ax.grid(True, alpha=0.25, linestyle='--', linewidth=1)
ax.set_facecolor('#f8f9fa')

# Add value labels on bars
for i, (bar, val) in enumerate(zip(bars, values)):
    width = bar.get_width()
    x_pos = width + (0.15 if width > 0 else -0.15)
    ha = 'left' if width > 0 else 'right'
    
    ax.text(x_pos, bar.get_y() + bar.get_height()/2.,
           f'{val:.3f}',
           ha=ha, va='center', 
           fontsize=12, fontweight='bold', color='#0d1b2a')

# Highlight Transmission Type with special border
if 'Transmission Type' in feature_names:
    trans_idx = feature_names.index('Transmission Type')
    bars[trans_idx].set_edgecolor('#FF6B35')
    bars[trans_idx].set_linewidth(4)

# Add model info
ax.text(0.98, 0.02, f'Model: Logistic Regression', 
       transform=ax.transAxes, fontsize=11, ha='right', va='bottom',
       style='italic', color='#666',
       bbox=dict(boxstyle='round,pad=0.4', facecolor='white', alpha=0.8))

plt.tight_layout()
plt.show()

print(f"\n{'='*80}")
print("VISUALIZATION COMPLETE")
print(f"{'='*80}")

================================================================================
LOGISTIC REGRESSION - FEATURE IMPORTANCE ANALYSIS
================================================================================

Dataset: 98,836 vehicles (after removing missing values)

Premium Threshold: $19,240
Premium vehicles: 29,654 (30.0%)
Non-Premium vehicles: 69,182 (70.0%)

Transmission encoding:
  Automatic: 0
  Manual: 1
  Other: 2
  Semi-Auto: 3

Fuel type encoding:
  Diesel: 0
  Electric: 1
  Hybrid: 2
  Other: 3
  Petrol: 4

Features: 7
Samples: 98,836

Training set: 79,068 samples
Test set: 19,768 samples

✓ Features standardized

--------------------------------------------------------------------------------
TRAINING LOGISTIC REGRESSION MODEL
--------------------------------------------------------------------------------

findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial

✓ Model trained successfully

Performance Metrics:
  Accuracy: 0.8937 (89.37%)
  ROC AUC: 0.9496

--------------------------------------------------------------------------------
FEATURE IMPORTANCE (LOGISTIC REGRESSION COEFFICIENTS)
--------------------------------------------------------------------------------

Rank   Feature                   Coefficient     Interpretation      
--------------------------------------------------------------------------------
1      engineSize                 +2.8079      Positive ↑
2      year                       +2.0089      Positive ↑
3      mileage                    -1.2424      Negative ↓
4      transmission_encoded       +0.3962      Positive ↑
5      fuelType_encoded           +0.2249      Positive ↑
6      tax                        +0.2058      Positive ↑
7      mpg                        +0.1098      Positive ↑
--------------------------------------------------------------------------------

findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial

================================================================================
VISUALIZATION COMPLETE
================================================================================

Engine Size & Year are the strongest premium drivers as larger engines (+2.291) and newer model years (+2.037) significantly increase premium likelihood—prioritize these attributes when acquiring inventory for premium positioning.

Mileage is the biggest negative factor as higher mileage (-1.288) strongly reduces premium classification—low-mileage vehicles retain value and should be priced accordingly.

Transmission type matters as Manual transmission (-0.620) decreases premium likelihood, while Semi-Auto (+0.477) and Automatic (+0.266) increase it—target automatic vehicles for premium buyers seeking convenience.

MPG has minimal impact as fuel efficiency (-0.010) barely influences premium status—premium buyers prioritize performance over economy.

Continuous Outcome Prediction

Downloading libraries

import polars as pl
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.pipeline import Pipeline
from sklearn.linear_model import LinearRegression
from sklearn.ensemble import RandomForestRegressor
from sklearn.svm import SVR
import plotly.express as px
import numpy as np

Target & Features

# Target variable
target = "price"

# Feature lists
categorical_features = ["Brand", "transmission", "fuelType"]
numeric_features = ["year", "mileage", "tax", "mpg", "engineSize"]

features = categorical_features + numeric_features

Train-test & Split(Continuous Outcome Prediction) (Polars to Pandas)

# Convert entire Polars DF into Pandas for the split
cars_pd = cars.to_pandas()

train_pd, test_pd = train_test_split(cars_pd, test_size=0.3, random_state=11109)

# Convert back to Polars
train_df = pl.DataFrame(train_pd)
test_df = pl.DataFrame(test_pd)

Preprocessing

preprocessor = ColumnTransformer(
    transformers=[
        ('num', StandardScaler(), numeric_features),
        ('cat', OneHotEncoder(handle_unknown='ignore'), categorical_features)
    ]
)

ML Pipelines (COP)

lr_pipeline = Pipeline([
    ('preprocessor', preprocessor),
    ('model', LinearRegression())
])

rf_pipeline = Pipeline([
    ('preprocessor', preprocessor),
    ('model', RandomForestRegressor(n_estimators=120, random_state=11109))
])

svm_pipeline = Pipeline([
    ('preprocessor', preprocessor),
    ('model', SVR())
])

Train and Predict (COP)

# Fit and predict
X_train = train_df.select(features).to_pandas()
y_train = train_df[target].to_pandas()

X_test = test_df.select(features).to_pandas()
y_test = test_df[target].to_pandas()

Linear Regression (Continous Outcome Prediction)

lr_pipeline.fit(X_train, y_train)

train_df = train_df.with_columns(
    pl.Series("lr_pred", lr_pipeline.predict(X_train))
)

test_df = test_df.with_columns(
    pl.Series("lr_pred", lr_pipeline.predict(X_test))
)

Random Forest (Continuous Outcome Prediction)

rf_pipeline.fit(X_train, y_train)

train_df = train_df.with_columns(
    pl.Series("rf_pred", rf_pipeline.predict(X_train))
)

test_df = test_df.with_columns(
    pl.Series("rf_pred", rf_pipeline.predict(X_test))
)

RMSE

def rmse(actual, pred):
    return np.sqrt(np.mean((actual - pred)**2))

rmse_lr = rmse(test_df[target].to_numpy(), test_df["lr_pred"].to_numpy())
rmse_rf = rmse(test_df[target].to_numpy(), test_df["rf_pred"].to_numpy())

print("RMSE Results:")
print(f"Linear Regression RMSE: {rmse_lr:.2f}")
print(f"Random Forest RMSE: {rmse_rf:.2f}")

RMSE Results:
Linear Regression RMSE: 4801.33
Random Forest RMSE: 2387.86

Visualization

# Simple scatter plot

def plot_pred(df, pred_col, title):
    fig = px.scatter(
        df,
        x=target,
        y=pred_col,
        title=title,
        labels={"price": "Actual Price", pred_col: "Predicted Price"}
    )
    fig.show()

plot_pred(test_df, "lr_pred", "Linear Regression: Actual vs Predicted Price")
plot_pred(test_df, "rf_pred", "Random Forest: Actual vs Predicted Price")

Interpretation of Visualizations

1.Linear Regression Plot:

The points are widely scattered and do not follow a strong straight-line pattern.

This means the linear regression model is struggling to capture the relationship between the input features and the actual bike price.

Many predictions are far from the actual values, with noticeable under- and over-prediction.

This indicates that the relationship in the data is non-linear, and linear regression is too simple for this dataset.

2.Random Forest Plot

The points lie much closer to a diagonal upward trend, meaning predictions follow actual prices more accurately.

The scatter is tighter, showing less error and better model fit compared to linear regression.

Random Forest captures complex nonlinear patterns in the dataset, which is why it performs significantly better.

Overall, Random Forest provides more reliable and stable predictions for bike price.

Brand Postioning Analysis

import statsmodels.api as sm
import pandas as pd

df = cars.to_pandas()

# Features to predict price
X = df[['year','mileage','engineSize','mpg','tax']]
X = sm.add_constant(X)
y = df['price']

model = sm.OLS(y, X).fit()

# Predicted (optimal) price
df["predicted_price"] = model.predict(X)

# Residual = actual - predicted
df["pricing_gap"] = df["price"] - df["predicted_price"]

Average Pricing Gap by Brand

brand_pricing = df.groupby("Brand")["pricing_gap"].mean().sort_values()
brand_pricing

px.bar(
    brand_pricing,
    title="Average Pricing Gap by Brand (Under/Overpricing Trend)",
    labels={"value": "Avg Pricing Gap", "index": "Brand"},
    color=brand_pricing.values
)

Brand Postioning Analysis

# Summary stats by brand
brand_summary = (
    cars.group_by("Brand")
        .agg([
            pl.mean("price").alias("avg_price"),
            pl.mean("mileage").alias("avg_mileage"),
            pl.mean("engineSize").alias("avg_engineSize"),
            pl.mean("mpg").alias("avg_mpg"),
            pl.len().alias("count")
        ])
        .sort("avg_price", descending=True)
)

brand_summary

shape: (9, 6)

Brand	avg_price	avg_mileage	avg_engineSize	avg_mpg	count
str	f64	f64	f64	f64	u32
"Mercedes"	24698.59692	21949.559037	2.07153	55.155843	13119
"Audi"	22896.685039	24827.244001	1.930709	50.770022	10668
"BMW"	22733.408867	25496.98655	2.167767	56.399035	10781
"VW"	16838.952365	22092.785644	1.600693	53.753355	15157
"Skoda"	14275.449338	20118.45205	1.433509	56.589165	6267
"Toyota"	12522.391066	22857.413921	1.471297	63.042223	6738
"Ford"	12279.756415	23363.630504	1.350827	57.906991	17965
"Hyundai"	12262.005988	22141.218674	1.446751	52.384209	4509
"Vauxhall"	10406.457893	23499.298636	1.417232	51.535007	13632

Brand Premium Score (Price vs Engine Size Adjusted)

import statsmodels.api as sm
import pandas as pd

df = cars.to_pandas()

X = df[['engineSize', 'mileage', 'year']]
X = sm.add_constant(X)
y = df['price']

model = sm.OLS(y, X).fit()
df["expected_price"] = model.predict(X)
df["premium_score"] = df["price"] - df["expected_price"]

brand_premium = df.groupby("Brand")["premium_score"].mean().sort_values(ascending=False)
brand_premium

Brand
Audi        3064.023872
Mercedes    2545.372264
VW           410.097296
BMW          168.600625
Ford        -398.537088
Skoda       -759.469816
Toyota     -1471.233709
Hyundai    -1928.945866
Vauxhall   -3197.122043
Name: premium_score, dtype: float64

Holistic Premium Score Model (Using ALL Features)

import statsmodels.api as sm
import pandas as pd

# Convert Polars → Pandas
df = cars.to_pandas()

# --- Features for predicting expected price ---
X = df[['engineSize', 'mileage', 'year', 'mpg', 'tax']]
X = sm.add_constant(X)

y = df['price']

# --- Fit OLS regression model ---
model = sm.OLS(y, X).fit()

# --- Predict expected fair value ---
df["expected_price"] = model.predict(X)

# --- Premium Score (Actual - Expected Price) ---
df["premium_score"] = df["price"] - df["expected_price"]

# --- Brand-level average premium score ---
brand_premium = (
    df.groupby("Brand")["premium_score"]
      .mean()
      .sort_values(ascending=False)
)

brand_premium

Brand
Audi        2976.221115
Mercedes    2634.557861
VW           334.512611
BMW          305.454716
Ford        -375.324828
Skoda       -748.748018
Toyota     -1331.554312
Hyundai    -2018.048664
Vauxhall   -3313.519487
Name: premium_score, dtype: float64

Brand Premium Score (Actual Price - Expected Price)

import statsmodels.api as sm
import pandas as pd
import plotly.express as px

# Convert Polars → Pandas
df = cars.to_pandas()

# Features for predicting expected price
X = df[['engineSize', 'mileage', 'year', 'mpg', 'tax']]
X = sm.add_constant(X)
y = df['price']

# Fit OLS regression
model = sm.OLS(y, X).fit()

# Compute premium score
df["expected_price"] = model.predict(X)
df["premium_score"] = df["price"] - df["expected_price"]

# Brand-level premium score
brand_premium = (
    df.groupby("Brand")["premium_score"]
      .mean()
      .sort_values(ascending=False)
      .reset_index()
)

# --- Premium Score Bar Chart ---
fig = px.bar(
    brand_premium,
    x="Brand",
    y="premium_score",
    title="Brand Premium Score (Actual Price vs Expected Price)",
    labels={"premium_score": "Avg Premium"},
    color="premium_score",
    color_continuous_scale="RdYlGn",   # Red → low premium, Green → high premium
    text="premium_score",
)

fig.update_traces(texttemplate='%{text:.0f}', textposition="outside")
fig.update_layout(xaxis_tickangle=45, height=500)
fig.show()

This chart compares each brand’s actual price with its expected price, generating a “premium score.” Audi and Mercedes command the highest positive premiums, indicating customers are willing to pay substantially more than expected, reflecting strong brand equity. VW and BMW show slight positive premiums, suggesting moderate brand strength. Ford and Skoda fall into mild negative territory, while Toyota, Hyundai, and especially Vauxhall have large negative premiums, meaning they sell for significantly less than expected. This signals weaker brand perception or strong price–value competition.

Marketing Implication:

Premium brands like Audi and Mercedes should continue leveraging their strong brand equity in messaging and maintain premium pricing strategies. VW and BMW can reinforce quality cues to further strengthen willingness to pay. Brands with negative premiums (Toyota, Hyundai, Vauxhall) should focus on improving perceived value through reliability messaging, product upgrades, or repositioning efforts. Vauxhall, with the deepest negative premium, may need rebranding or pricing adjustments to rebuild consumer trust and competitiveness.

Brand Quadrant Analysis (Strategic Positioning)

import plotly.graph_objects as go

# Calculate medians for quadrant lines
median_price = brand_summary["avg_price"].median()
median_mpg = brand_summary["avg_mpg"].median()

fig = go.Figure()

# Add scatter points with improved colors
fig.add_trace(go.Scatter(
    x=brand_summary["avg_mpg"],
    y=brand_summary["avg_price"],
    mode="markers+text",
    text=brand_summary["Brand"],
    textposition="top center",
    marker=dict(
        size=brand_summary["count"] / brand_summary["count"].max() * 50 + 15,
        color=brand_summary["avg_price"],
        colorscale="Bluyl",
        showscale=True,
        colorbar=dict(title="Price ($)", tickprefix="$", tickformat=",.0f"),
        line=dict(width=1, color="white")
    ),
    textfont=dict(size=11, color="#333", family="Arial")
))

# Add quadrant lines
fig.add_hline(y=median_price, line_dash="dash", line_color="#888", line_width=2)
fig.add_vline(x=median_mpg, line_dash="dash", line_color="#888", line_width=2)

# Calculate positions for labels (corners of the chart to avoid overlap)
x_min = brand_summary["avg_mpg"].min()
x_max = brand_summary["avg_mpg"].max()
y_min = brand_summary["avg_price"].min()
y_max = brand_summary["avg_price"].max()

# PREMIUM PERFORMANCE (top left corner)
fig.add_annotation(
    x=x_min + 1, 
    y=y_max + 1500,
    text="<b>PREMIUM PERFORMANCE</b>", 
    showarrow=False, 
    font=dict(size=11, color="black"),
    bgcolor="rgba(41, 128, 185, 0.15)",
    borderpad=6
)

# PREMIUM EFFICIENT (top right corner)
fig.add_annotation(
    x=x_max - 1, 
    y=y_max + 1500,
    text="<b>PREMIUM EFFICIENT</b>", 
    showarrow=False, 
    font=dict(size=11, color="black"),
    bgcolor="rgba(39, 174, 96, 0.15)",
    borderpad=6
)

# VALUE PERFORMANCE (bottom left corner)
fig.add_annotation(
    x=x_min + 1, 
    y=y_min - 1000,
    text="<b>VALUE PERFORMANCE</b>", 
    showarrow=False, 
    font=dict(size=11, color="black"),
    bgcolor="rgba(230, 126, 34, 0.15)",
    borderpad=6
)

# VALUE EFFICIENT (bottom right corner)
fig.add_annotation(
    x=x_max - 1, 
    y=y_min - 1000,
    text="<b>VALUE EFFICIENT</b>", 
    showarrow=False, 
    font=dict(size=11, color="black"),
    bgcolor="rgba(26, 188, 156, 0.15)",
    borderpad=6
)

fig.update_layout(
    title=dict(
        text="<b>Brand Strategic Quadrant Analysis</b><br><sup>Price vs Fuel Efficiency | Bubble Size = Market Share</sup>",
        font=dict(size=18, color="#2c3e50"),
        x=0.5
    ),
    xaxis_title="Average MPG (Fuel Efficiency)",
    yaxis_title="Average Price ($)",
    height=650,
    width=1000,
    xaxis=dict(
        gridcolor="#eee", 
        tickfont=dict(size=11),
        range=[x_min - 3, x_max + 3]
    ),
    yaxis=dict(
        tickprefix="$", 
        tickformat=",.0f", 
        gridcolor="#eee", 
        tickfont=dict(size=11),
        range=[y_min - 2500, y_max + 3000]
    ),
    plot_bgcolor="white",
    paper_bgcolor="white",
    showlegend=False
)

fig.show()

This map compares average price versus average mileage for major car brands. BMW, Audi, and Mercedes remain positioned as premium brands, offering higher mileage but at significantly higher prices. VW sits in the mid-range with moderate pricing and mileage. Brands like Toyota, Hyundai, Ford, and Vauxhall cluster in the affordable segment, offering competitive mileage at lower price points. Skoda shows relatively lower mileage for its price, indicating a value–perception gap.

Marketing Implication:

Luxury brands should emphasize their balance of performance, comfort, and mileage to reinforce premium value. Mid-market brands like VW can market themselves as practical upgrades that offer strong mileage at reasonable prices. Budget brands (Toyota, Hyundai, Ford, Vauxhall) should highlight fuel efficiency and long-term cost savings for value-conscious buyers. Skoda may need to improve its mileage perception or adjust pricing to stay competitive in the value-driven segment.

This brand-positioning map compares average price vs. average engine size across car brands. BMW, Mercedes, and Audi occupy the premium segment, offering larger engines at significantly higher prices. Mid-market brands like VW sit in the center with moderate pricing and engine size. Meanwhile, Ford, Toyota, Hyundai, Skoda, and Vauxhall cluster in the affordable, smaller-engine segment, appealing to value-focused consumers. Overall, the chart highlights a clear separation between luxury and mass-market brands based on engine performance and price.

Marketing Implication:

The map highlights a clear premium vs. value split. Luxury brands (BMW, Mercedes, Audi) should continue emphasizing performance, engineering quality, and prestige to justify their higher prices. Mid-range brands like VW can position themselves as balanced choices combining reliability and moderate performance. Budget-focused brands (Toyota, Hyundai, Ford, Vauxhall) should reinforce value, fuel efficiency, and affordability. This segmentation helps brands tailor messaging, product strategy, and competitive positioning to their target consumer groups.

Pricing Optimization

Price Correlation Matrix

numeric_cols = ["year", "price", "mileage", "tax", "mpg", "engineSize"]
corr_matrix = cars.select(numeric_cols).to_pandas().corr()

fig = px.imshow(
    corr_matrix,
    text_auto=".2f",
    color_continuous_scale="RdYlBu_r",
    title="Price Correlation Matrix: What Drives Vehicle Price?",
    labels=dict(color="Correlation"),
    x=numeric_cols,
    y=numeric_cols,
    zmin=-1,
    zmax=1
)

fig.update_layout(
    height=600,
    width=700,
    title=dict(
        text="<b>Price Correlation Matrix: What Drives Vehicle Price?</b>",
        font=dict(size=18),
        x=0.5,
        xanchor="center"
    ),
    font=dict(family="Arial", size=12),
    coloraxis_colorbar=dict(
        title="Correlation",
        tickvals=[-1, -0.5, 0, 0.5, 1],
        len=0.8,
        thickness=15
    ),
    margin=dict(l=80, r=50, t=80, b=50)
)

fig.update_traces(
    textfont=dict(size=12, color="black"),
    hovertemplate="<b>%{x}</b> vs <b>%{y}</b><br>Correlation: %{z:.2f}<extra></extra>"
)

fig.update_xaxes(tickangle=45, tickfont=dict(size=11))
fig.update_yaxes(tickfont=dict(size=11))

fig.show()

Visualizations (Pricing Optimization)

import polars as pl
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Set style for better presentation
sns.set_style("whitegrid")
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['font.sans-serif'] = ['Arial']

# Convert to pandas for analysis
df = cars.to_pandas()

# Create the comprehensive price sensitivity dashboard
fig = plt.figure(figsize=(20, 14))
gs = fig.add_gridspec(2, 3, hspace=0.35, wspace=0.35)

# Main title with better styling
fig.suptitle('COMPREHENSIVE PRICE SENSITIVITY ANALYSIS\nMarket Wheels - Understanding Price Drivers',
             fontsize=26, fontweight='bold', y=0.98, color='#1a1a2e')

# Sample data for faster plotting if dataset is large
if len(df) > 10000:
    sample_df = df.sample(n=5000, random_state=42)
else:
    sample_df = df

# Define professional color scheme
primary_blue = '#4361EE'
trend_red = '#EF233C'
bar_color_1 = '#06D6A0'
bar_color_2 = '#4CC9F0'


# PLOT 1: Price Sensitivity to Mileage
ax1 = fig.add_subplot(gs[0, 0])
ax1.scatter(sample_df['mileage'], sample_df['price'], s=15, alpha=0.5,
           c=primary_blue, edgecolors='white', linewidth=0.3)

# Fit trend line
z1 = np.polyfit(df['mileage'], df['price'], 1)
p1 = np.poly1d(z1)
x_line1 = np.linspace(df['mileage'].min(), df['mileage'].max(), 100)
ax1.plot(x_line1, p1(x_line1), color=trend_red, linestyle='--', linewidth=3, alpha=0.85)

ax1.set_xlabel('Mileage (miles)', fontsize=13, fontweight='bold', color='#2d3436')
ax1.set_ylabel('Price ($)', fontsize=13, fontweight='bold', color='#2d3436')
ax1.set_title('Price Sensitivity to Mileage', fontsize=15, fontweight='bold',
             pad=20, color='#1a1a2e')
ax1.grid(True, alpha=0.25, linestyle='--', linewidth=0.8)
ax1.set_facecolor('#f8f9fa')

# Add correlation with better styling
corr_mileage = df['mileage'].corr(df['price'])
ax1.text(0.05, 0.95, f'Correlation: {corr_mileage:.3f}',
         transform=ax1.transAxes, fontsize=12, verticalalignment='top',
         fontweight='bold', color='#2d3436',
         bbox=dict(boxstyle='round,pad=0.6', facecolor='#fffbeb',
                  edgecolor='#fbbf24', linewidth=2, alpha=0.9))

# PLOT 2: Price Sensitivity to Vehicle Age
ax2 = fig.add_subplot(gs[0, 1])
ax2.scatter(sample_df['year'], sample_df['price'], s=15, alpha=0.5,
           c=primary_blue, edgecolors='white', linewidth=0.3)

# Fit trend line
z2 = np.polyfit(df['year'], df['price'], 1)
p2 = np.poly1d(z2)
x_line2 = np.linspace(df['year'].min(), df['year'].max(), 100)
ax2.plot(x_line2, p2(x_line2), color=trend_red, linestyle='--', linewidth=3, alpha=0.85)

ax2.set_xlabel('Year', fontsize=13, fontweight='bold', color='#2d3436')
ax2.set_ylabel('Price ($)', fontsize=13, fontweight='bold', color='#2d3436')
ax2.set_title('Price Sensitivity to Vehicle Age', fontsize=15, fontweight='bold',
             pad=20, color='#1a1a2e')
ax2.grid(True, alpha=0.25, linestyle='--', linewidth=0.8)
ax2.set_facecolor('#f8f9fa')

# Add correlation
corr_year = df['year'].corr(df['price'])
ax2.text(0.05, 0.95, f'Correlation: {corr_year:.3f}',
         transform=ax2.transAxes, fontsize=12, verticalalignment='top',
         fontweight='bold', color='#2d3436',
         bbox=dict(boxstyle='round,pad=0.6', facecolor='#fffbeb',
                  edgecolor='#fbbf24', linewidth=2, alpha=0.9))


# PLOT 3: Price Sensitivity to Engine Size
ax3 = fig.add_subplot(gs[0, 2])
ax3.scatter(sample_df['engineSize'], sample_df['price'], s=15, alpha=0.5,
           c=primary_blue, edgecolors='white', linewidth=0.3)

# Fit trend line
z3 = np.polyfit(df['engineSize'], df['price'], 1)
p3 = np.poly1d(z3)
x_line3 = np.linspace(df['engineSize'].min(), df['engineSize'].max(), 100)
ax3.plot(x_line3, p3(x_line3), color=trend_red, linestyle='--', linewidth=3, alpha=0.85)

ax3.set_xlabel('Engine Size (Liters)', fontsize=13, fontweight='bold', color='#2d3436')
ax3.set_ylabel('Price ($)', fontsize=13, fontweight='bold', color='#2d3436')
ax3.set_title('Price Sensitivity to Engine Size', fontsize=15, fontweight='bold',
             pad=20, color='#1a1a2e')
ax3.grid(True, alpha=0.25, linestyle='--', linewidth=0.8)
ax3.set_facecolor('#f8f9fa')

# Add correlation
corr_engine = df['engineSize'].corr(df['price'])
ax3.text(0.05, 0.95, f'Correlation: {corr_engine:.3f}',
         transform=ax3.transAxes, fontsize=12, verticalalignment='top',
         fontweight='bold', color='#2d3436',
         bbox=dict(boxstyle='round,pad=0.6', facecolor='#fffbeb',
                  edgecolor='#fbbf24', linewidth=2, alpha=0.9))


# PLOT 4: Average Price by Transmission Type (VERTICAL BARS)
ax4 = fig.add_subplot(gs[1, 0])
trans_data = df.groupby('transmission')['price'].mean().sort_values(ascending=False)

# Create gradient colors
colors_trans = plt.cm.viridis(np.linspace(0.3, 0.8, len(trans_data)))
bars4 = ax4.bar(range(len(trans_data)), trans_data.values,
                color=colors_trans, alpha=0.85, edgecolor='#2d3436', linewidth=2.5)

ax4.set_xticks(range(len(trans_data)))
ax4.set_xticklabels(trans_data.index, rotation=0, ha='center', fontsize=11, fontweight='bold')
ax4.set_ylabel('Average Price ($)', fontsize=13, fontweight='bold', color='#2d3436')
ax4.set_title('Average Price by Transmission Type', fontsize=15, fontweight='bold',
             pad=20, color='#1a1a2e')
ax4.grid(True, alpha=0.25, axis='y', linestyle='--', linewidth=0.8)
ax4.set_facecolor('#f8f9fa')

# Add value labels on bars with better styling
for i, bar in enumerate(bars4):
    height = bar.get_height()
    ax4.text(bar.get_x() + bar.get_width()/2., height,
             f'${height:,.0f}',
             ha='center', va='bottom', fontsize=11, fontweight='bold',
             color='#2d3436', bbox=dict(boxstyle='round,pad=0.3',
                                       facecolor='white', alpha=0.8))

# Set same y-axis range as fuel type plot (will be calculated after fuel plot)
trans_max = trans_data.max()



# PLOT 5: Average Price by Fuel Type (VERTICAL BARS)
ax5 = fig.add_subplot(gs[1, 1])
fuel_data = df.groupby('fuelType')['price'].mean().sort_values(ascending=False)

# Create gradient colors
colors_fuel = plt.cm.plasma(np.linspace(0.2, 0.8, len(fuel_data)))
bars5 = ax5.bar(range(len(fuel_data)), fuel_data.values,
                color=colors_fuel, alpha=0.85, edgecolor='#2d3436', linewidth=2.5)

ax5.set_xticks(range(len(fuel_data)))
ax5.set_xticklabels(fuel_data.index, rotation=45, ha='right', fontsize=11, fontweight='bold')
ax5.set_ylabel('Average Price ($)', fontsize=13, fontweight='bold', color='#2d3436')
ax5.set_title('Average Price by Fuel Type', fontsize=15, fontweight='bold',
             pad=20, color='#1a1a2e')
ax5.grid(True, alpha=0.25, axis='y', linestyle='--', linewidth=0.8)
ax5.set_facecolor('#f8f9fa')

# Add value labels on bars
for i, bar in enumerate(bars5):
    height = bar.get_height()
    ax5.text(bar.get_x() + bar.get_width()/2., height,
             f'${height:,.0f}',
             ha='center', va='bottom', fontsize=11, fontweight='bold',
             color='#2d3436', bbox=dict(boxstyle='round,pad=0.3',
                                       facecolor='white', alpha=0.8))

# Set same Y-axis scale for both bar charts for better comparison

fuel_max = fuel_data.max()
y_max = max(trans_max, fuel_max) * 1.15  # Add 15% padding
ax4.set_ylim(0, y_max)
ax5.set_ylim(0, y_max)


plt.tight_layout()
plt.show()

findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial
findfont: Generic family 'sans-serif' not found because none of the following families were found: Arial

Recommendations and Implications

1.Rebalance Inventory Using Price Elasticity

Recommendation:

Shift inventory toward mid-range and economy vehicles, as these segments show the highest demand and offer steady margins. Keep premium models in smaller quantities but price them higher to maintain exclusivity.

Implication:

This approach improves overall turnover, reduces holding costs, and ensures the dealership stocks cars that align with real market demand patterns.

2.Apply Premium Pricing for Engine Size & Automatic Transmissions

Recommendation:

Introduce consistent pricing rules such as adding a fixed premium for larger engines (e.g., +$1,000 per 0.5L) and pricing automatic variants 8–12% higher across the board.

Implication:

Feature-based pricing improves transparency, strengthens customer trust, and increases margins by charging appropriately for in-demand specifications.

3.Use Depreciation-Adjusted Stock Rotation for Aging Vehicles

Recommendation:

Move older or high-mileage inventory faster by using timed discounts and offering service bundles for slow-moving models.

Implication:

This minimizes depreciation losses, improves cash flow, and keeps the dealership’s inventory fresh and competitive.

4.Use Customer Segmentation for Targeted Marketing & Reduced Negotiation

Recommendation:

Customer segmentation can be used to customize marketing and price by discriminating between value purchasers, who prioritize high-MPG, low-maintenance, inexpensive automobiles, and performance buyers, who prefer high-engine, lower-MPG, luxury or performance models. Position efficient automobiles as “Low Ownership Cost” options, and higher-engine vehicles as premium or performance options, tailoring promotions and communication to each consumer profile.

Implication:

This personalized strategy eliminates negotiation friction, promotes customer happiness, and boosts conversion by ensuring buyers are quickly presented with vehicles that match their goals and budget. It improves marketing clarity and makes the dealership’s product offering appear more relevant and tailored to each consumer segment.

Final Recommendation:

Customer segmentation can be used to customize marketing and price by discriminating between value purchasers, who prioritize high-MPG, low-maintenance, inexpensive automobiles, and performance buyers, who prefer high-engine, lower-MPG, luxury or performance models. Position efficient automobiles as “Low Ownership Cost” options, and higher-engine vehicles as premium or performance options, tailoring promotions and communication to each consumer profile.

Final Implication:

This targeted approach reduces negotiation friction, improves customer satisfaction, and increases conversion by ensuring buyers immediately encounter vehicles that match their priorities and budget. It strengthens marketing clarity and makes the dealership’s product offering feel more relevant and personalised to each customer segment.

Conclusion

According to the data, pricing, customer demand, and brand impression are all influenced by a small number of essential vehicle features, namely mileage, year, engine size, MPG, and gearbox. Using these insights, the dealership can transition from intuition-based decisions to a fully data-driven strategy that establishes optimal prices, targets the appropriate consumer categories, and presents each brand successfully.

Overall, the findings conclude that data-driven pricing and segmentation create the strongest path to higher margins, faster inventory turnover, and more relevant marketing in the competitive used-car market.